Research methodology: guide to process tracing and realist evaluation

Process tracing was first trialled as a new approach within a series of theory-based evaluations of Christian Aid’s governance portfolio in 2015. The aim was specifically to understand and evidence how Christian Aid and its partners’ accountability practices were contributing to building more (downwardly) accountable relationships more broadly in the context of Bangladesh.

The evaluation team added elements of realist evaluation to their evaluation design, in order to allow for greater flexibility of the methodology and ‘user-friendliness’ of the findings.

‘Flexibility’ because selection of methodology in this case preceded the final agreement on evaluation questions, and combining the two different methodologies would provide more room for the evaluators to adapt and tailor the approach based on the eventually defined questions; greater ‘user-friendliness’ due to realist evaluation’s focus on potentially more actionable ‘what works where for whom?’ questions that would complement the more theoretical ‘tracing’ of competing explanations through process tracing. 

This guide focusses primarily on the steps necessary to conduct process tracing but includes a brief section on how and where to combine this with realist evaluation. There is a list of selected references at the end of the document for those interested in further reading.